INDEX

Explanations

elements related to programming or coding structures and commands

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 NSCoder

-0.84

transQ

-0.77

 gynhyrchwyd

-0.68



-0.68

 disambiguazione

-0.65

IVEREF

-0.60

 <=",

-0.60

aarrggbb

-0.60

ivably

-0.59

StoryboardSegue

-0.59

POSITIVE LOGITS

الحياه

0.43

étation

0.36

QueryResult

0.33

terung

0.30

PointerException

0.30

 velkommen

0.30

 Feier

0.29

 interpreted

0.29

Conclusiones

0.29

 dahin

0.29

Activations Density 0.133%

elements related to programming or coding structures and commands

No Comments

No Known Activations